Name: The Suite Customer Forum
Uploaded: 2026-04-29T17:01:00.894Z
Duration: 1 h 8 min 4 s
Description: The Suite Customer Forum

Transcript for "The Suite Customer Forum": Good morning, good afternoon, good evening, depending on where you're calling in from the world. My name is David Lee. I am a program, customer program manager here at Precisely, and we're really excited to have you join us today for our first Suite Customer Forum. We have a jam packed agenda today for you. So let's get started. Welcome. You will see we are on our new platform Goldcast. There are a couple of ways you can interact with us. On the top right hand side, you will see a chat box. You'll see that I dropped in hello from Fort Myers, Florida. That's where I'm calling in from. If you wouldn't mind dropping in where you're calling in from, we'd love to know where you're at. This is where you can chat with your colleagues, chat with other attendees, and ask questions to the audience. In the q and a section, you can ask questions of our presenters, and you'll see that on the top, button as well. And later on during our program, we'll have a chance from, our audience to raise your hand, join us on stage and ask a question of our experts toward the end of the program. So we we look forward to having you join us and hear your voice today. So welcome. We're really excited to have you join us. And as we get started, we have a jam packed agenda today. We'll hear from Antonio, welcoming us. Antonio, if you wouldn't mind joining us on the stage, then Dylan Conrad will talk about spatial enrichment and the APIs, and then Nina will talk about shifting from search using AI assistant and developer portal. Dave Mosher will talk about some stories in the field about how people are using the data integrity suite. And finally, what's not on the slide is we'll have a q and a session at the very end and then stay tuned for some announcements. And it's my pleasure to introduce to you Antonio, who is director of product marketing. Antonio, welcome. Welcome, everybody. It's so good to to see you all here, today. As as David had mentioned, my name is Antonio Caturno, and I'm one of our directors of product marketing, here at at Precisely. I I support the the the data integrity, suite is one of the primary products that I I support here at Precisely, and I'm I'm looking forward to to chatting with you all, today. I know that we have an initial poll here that we're gonna be, launching just to try to understand a little bit more from you all and and what you're interested in in learning, from us. So I'll I'll, I'll let David and and team kick kick that off before I get. going. If you on the top right, you will see a poll, question. So what are you hoping to learn today? Are you hoping to learn more about the data integrity suite, more about the road map and services, tips and tricks, visions? You can select one or all of the options here. And for our friends that are, precisely employees, we ask that you let the customers, answer that poll. So if you could just test that poll, that would be great. Take that poll. Not test, that poll. We we are sort of in the testing unit. This is a new platform for us, so we're we're really excited. sure. Absolutely. While we start to get those answers, Antonio, I I think we can start your trek. So. feel free. to take that poll anytime. No doubt. Alright. Again, thank you all for being here. This is fantastic. It's it's something that's sort of been long time long, long time coming. So but before we sort of kinda jump into the, broader conversation and we get, like, kinda Dylan and Nina and Dave, involved, I just wanted to, like, really quickly spend a few minutes on, like, why we're here. Like, why did we do finally decide to to do this? And so, like, for a while now, we've sort of been paying close attention to what our users have been telling us across various channels. This could be things that, you all are telling us through support conversations or customer success conversations, when we meet at, at conferences or in in in one on one, get togethers. And, like, throughout all of those various, conversations, we've been getting a lot of similar signals, from you all. So now click, you know, you could think that each of those various, signals are are sort of, you know, points of data, points in in time, but they're sort of across the same, exact, theme. And, you know, when you start to get so many of these things across the various customer conversations, sort of have to finally say, you know what, we need to do something about this. We need to start building a little bit of a community. The two major things that came out of taking a look at all those various conversations, gathering all of that feedback was essentially that, you know, I know someone out there has already solved this problem and I just I can't I can't find them. You know, whether it's, you know, developers that we speak to, builders within within the suite are hitting these these challenges and then, you know, trying to figure out clever solutions to those and know that that knowledge was was getting, sort of locked into perhaps individual teams or individuals across, other organizations. No real great way to be able to share that back out. Or maybe I came up with a novel solution to a challenge that I think everybody else would really get a lot of value from. It wasn't a great way to be able to share that with other folks. And then the second one, which is really just the flip side of that, is that, you know, I'd love to know what other people are building and what other people, you know, other solutions that they that they found, what patterns they found that have worked at, at at various, scales. So these are like sort of those two things that kept, coming up. So it's pretty clear, you know, what was missing wasn't necessarily more documentation or another webinar, another blog post, or like something else like that. It was each other. And that was sort of the big piece that was sort of missing a space where people could, be able to share their work, be able to make connections with one another, and be able to find common solutions that we could bring forward. And that's really what today is about and what this this, Suite Customer Forum is going to be about, moving forward. So this isn't going to be something that's, you know, this isn't just gonna be another webinar. We don't wanna talk at you for for for an hour. I think here in this first session, there may be a little bit more that we're gonna present, but it's mostly just to kind of get the conversation and, started and really begin to turn this over to you all as a as a shared space and and start of a of a of a community and a place where developers and builders who are working on, you know, similar things and working with the with the same tools, be able to share what they're working on, ask common questions, challenging questions that that you may have, and then learn from learn from one another. And I know there's a good number of us, here today, and every one of you is here because you're actively working on our on our platform in the Data Integrity Suite. That's a rare space to be. I had the opportunity over the last year to do a few of these with some of our other products and with customers across the board, and that's been incredibly valuable for me as someone that isn't able to get out and to meet, to meet customers all that often. So this has been incredibly valuable. So I'm I'm very excited about being able to do this, with you all. So my hope is that by the end of today, you know a few more of the people that are in this with with you And then, you know, also be able to understand that you have a place to go to be able to ask these types of questions and and be a part of this, community. We're gonna be doing this more regularly. Today is the first one. So, like, as David had had launched in that initial poll, yes. We wanna know. what you wanna hear about today. We also want your feedback on this on this format, the the the topics that we're gonna be, covering, and what you wanna hear more about in the future. And that'll help shape, these types of of sessions, as as we, as we move forward. And why don't we look at that poll, Antonio? Amanda, if you could push the results there. No no surprise road map and was the was the top vote getter followed about more about the suite overall. So this is great. Thank you very much. Alright. Thank you. Good to know. Alright. And as, David had mentioned at the top, but I just just to kinda, reiterate, this is kinda what we have lined up for today. First, we're gonna walk through, MCP server. I mean, if you've been watching the the space of late, you know that MCP is sort of changing how developers and builders are connecting APIs and data to to AI. And and what we're shipping will meaningfully, like, expand what we're able to build on top of, the data integrity suite. Second, we're gonna, get into some of the enhancements we are making in the developer portal, including the new AI assistant that's built into it. And the goal there is really just to talk about, hey. Hey. I have an idea, and I want, I wanna get to kinda working code much faster with with with less, with less friction. So we'll kinda get into that there. And then third, we're gonna go over, to, today. We'll talk, a little bit about, you know, what you're building, swap notes, you know, open up for questions, discussion, and that's something I'm very excited about hearing from you all. So that's where we're headed. Thanks again for being here and for being a part of this first session that we're doing, with the the Data Integrity Suite customer forum. And, you know, it really wouldn't work without you being here and being a part of this community. So, again, thank you so much. And, with that, I'm gonna turn it over to Dylan. He's gonna be talking about APIs and and MCP. So thank you all. Looking forward to the conversation. Hello, folks. Thanks, Antonio. Can everyone hear me okay, David? Yep. All good. Alright. Excellent. So jumping right in here, a a majority of the members on this form, I expect, will be users of the precisely geo addressing APIs and geocoding capability, just from looking at the companies on the registration. But, the fact that almost any business, relies on accurate address data in some capacity. I'm sure there will be many discussions going forward on best practices for verifying, cleansing, and resolving address data. But once addresses are corrected, that's usually just part of the business need for that data. Often, data enrichment is required to better understand the locations at those addresses. At Precisely, we refer to this best practice related to addresses as correct and connect. Precisely's Datagraph, API is built to offer data enrichment seamlessly with our geo geocoding services so that you can connect data. So today, I wanted to provide a quick update on the latest enhancements to the Datagraph API, compare Datagraph to the precisely spatial APIs, and look at how we're working with clients to use MCP server to connect AI models to the to the, data integrity suite and drive actional reasoning with trusted data and not just web based hallucinations. So I'm gonna hop ahead here. Sorry. Just one configuration, thing quickly. Oh, adjust my camera a little bit. I feel like I'm a bit dark here for y'all. So looking at data graph, for those of you, that are unfamiliar, the DataGraph API is an API that we developed to expose the depth and breadth of the precisely precisely data portfolio through a single endpoint. It's a customizable API in that you can query a single data set or multiple data sets in a single query and return only the attributes required for a specific use case. We've, based it on our connected ID infrastructure, an ID based relationship system linking unique and persistent IDs in one dataset to those IDs in another dataset. And the foundation of this relationship system is the Precisely ID, our unique identifier for an address, which allows easy integration with our geocoding and geocoding workflows. Data graph is built specifically to support a wide range of data enrichment use cases, eliminating the need for spatial processing, complex joins, and data manipulation, all serviced through a single API. In addition to the Data Graph API, we also have a set of spatial APIs to support visualization, mapping, and spatial analytics types of use cases. These spatial services consist of mapping APIs, including raster tiles API, dynamic maps, and feature service, all built to OGC specifications. Previously, these were referred to by their OGC spec definitions, WMTS, WMS, WFS APIs, respectively. These are your more traditional GIS APIs, meaning they can easily be integrated into GIS software, such as MapInfo or ArcGIS or QGIS. But they can also be used to power web mapping applications. For instance, they can be used or for instance, Map Explorer on the precisely data experience is powered by our precisely spatial APIs. Or they can be used to embed mapping and visualization into dynamic reports and AI applications and tools. We also, have APIs that perform specific spatial analysis, such as summarize, search nearby, search at location, and overlap analysis, allowing clients to investigate relationships between their data, other third party datasets, and precisely data without requiring software and an environment to do that intense processing. Here, we have a comparison of the data products available, through both these different types of APIs, the data graph and the spatial APIs. For the most part, we look to expose the same data products through both technologies, to meet better to meet use cases, better served by each API's distinct capability. But being that, the APIs are built to support different use cases, there are naturally some discrepancies in the data available between the two. And that's primarily products that are built entirely for ID based enrichment, such as our data link program are available only through the data graph. Similarly, if a dataset is primarily exposed for visualization, we may only expose it via the spatial mapping APIs. But if you look here, you'll see that we have a wide range of data available. We have our complete address and property portfolio. We have demographics, neighborhoods, schools, business places, and a number of other datasets to support a wide range of use cases. Jumping ahead to the next slide, a quick look at the data graph road map here and recent enhancements that have been added. I'm not gonna go through every item on this, but some key additions and coming features to highlight are, first of all, Precisely's address geocode service is now available through the Datagraph API. So clients, with geocoding entitlement now have the option to use the standard REST API, or they can just use Datagraph directly if they're going to be using Datagraph for additional data enrichment anyways. As far as new datasets, we now have over 25 data products available through Data Graph, now that we've added Datalink for GeoX, property insights extracted from aerial imagery, are now available through our strategic partnership with GeoX. We've also added bulk Datagraph API, a new endpoint where a file can be submitted to the API for data enrichment, with one or more of the Datagraph datasets and then processed at scale. This is built on the same framework as our bulk geo addressing, API for those familiar with that. And then, coming in q two, we're working on a user interface for a Datagraph on the suite so that a user, potentially a nontechnical user even, can load a CSV and select data and attributes, that they require for enrichment, based on the data graph's capability, with no need for coding or any development resources. So we're really excited about that coming. Of course, we'll continue to expand the data available and query methods, to interact with that data through Datagraph as well. But a real key initiative in the works and that we're focusing effort on is our MCP server integration, and AI applications and tools, leveraging the Precisely APIs and exposing these as alternate ways to access Precisely data through the suite. And this is really what I wanted to, just show a few quick examples of, how precisely APIs can be used with MCP server, and how powerful when used with AI integrations and apps that precisely data can be in combination with the MCP server. And then we're really curious to hear, any use cases or discussion around MCP server and AI as, the folks on this forum are using within their own integrations. Dave, did I hear a question or or anything before I move. forward? Dylan, this is a good time actually. Amanda, if you could push the second poll, we do have a quick poll. When Dylan says MCP server, do you know what that means? Yes or no? Or kind of. As we get through here, drop your questions in again at any time and if you are brave and you want to join us on stage and ask a question directly, raise your hand and Amanda will promote you and we'll take your question live. Alright. Should we. wait? I can ramble on while we're waiting for the poll if you want, David. can ramble as much as you like, Dylan. Alright. So, off the switch, anytime. is It looks like we're about about half of the folks, a little more than half know what MCP is. So okay. Excellent. I assumed that it might be the case. So I'm gonna hop in first and just show a couple, quick, kinda simple examples. I'll be using Claude, which I find provides a stronger reasoning and a bit more practical for data flow, related business use cases. I would imagine that most of us in this forum are beginning to adopt AI tools like Cloud, ChatGPT, Copilot, within our companies. And these tools are becoming embedded in everyday work routines in in some way or fashion. These LLMs are great. They're a huge time saver, and powerful reasoners. But they, out of the box, they lack trusted, consistent responses with verified data. And without constraints and, instructing the agents, what data to use specifically, they can hallucinate, and they fabricate data, data points. So that's why there's so much buzz about MCP server is it allows you to basically put guardrails on AI agents and LL items. So with that, I'm going to just use my screen here a tiny bit and find the share button. One second. I feel like it moved on me. David? top left, there you go. yep. 3. Yeah. Perfect. Apologies. New platform as David was saying. So with that, good for a window. And while that comes up, Interesting. Dylan did have a question. You mentioned that you're gonna use Claude. We don't the question is, do I need Claude to access MTP? But I I believe you said it can be any LLM for You do not need, cloud to access m MCP. It can be used, with chat GPT. It it can be used with, Copilot and others. And we'll also be building AI tools and apps that leverage MCP under the hood, if you will, where you you wouldn't need to use any AI, agent directly. Claude is, one of the, preferred, I would say, in, you know, that folks are using right now. So here, I'm gonna share my screen. Apologize. I just had to adjust. It wasn't letting me show one half at a time. Can everyone see my screen here, David? Yep. Good to go. Alright. Excellent. So, what I I just wanna show a couple quick examples of, you know, using Claude just based on a a basic web search to start. So here I have a web search turned on, and I'm just gonna assure I have my connectors turned off. And I'm gonna ask Claude a simple question. With three, two, Denver. So live, Claude demo here, so wish me luck. So as we're all familiar, these AI tools are, like, amazing and and powerful and and give you a a very interesting response to almost any question that you can picture asking. But when you're acquiring it just on the LLM in a web search, the answers can be very inconsistent, can return, you know, inaccurate information, and so forth. So this is a very detailed response. But the if I were to ask a different location, a a different, area in the country, like, you're at a previous similar search for this address, I get some different sources that are used from the web for each of these responses, different format of the response, and so forth. It can really vary, specifically. To control and harness the power of AI with, trusted data, we can use MCP server to, I'll go to a new chat, turn off web search, and turn on my connector. Just a simple connect MTP, connection that I have for the data graph. I'm gonna go in manage skills and turn on, just one skill in our production configuration that we have. Dozens, hundreds of skills are being created, and I'm gonna ask the same question. Normally, I just talked to Claude, but I won't make you folks listen to me talking to an AI agent. So here, I'm querying, Claude using the MPC server and a specific skill set to first geocode or verify validate this address that I entered. I didn't include a ZIP code, so it's gonna validate it down to the ZIP code level. It's gonna geocode and append a precisely ID and the latitude longitude coordinates for that address. And then it's taking the precise the ID and running against nine separate queries in parallel against the Datagraph API, to return risk and property details for this location. Now, again, this is just a simple configuration that I put together specifically for this demo to compare web versus. It's not entirely optimized. Could have just queried all of these datasets in a single query against Datagraph and had a custom query set up as an action. They were queried as nine parallel queries in this one. And here I return, very specific detailed analysis about this property. I have the address, information, verified and completed to the zip four level. At the lat, long, and the precisely ID from the geocode. And then I've gone and enriched this location with the entire risk portfolio datasets available through, the data graphs. So we have the flood zone determination. No no coastal risk in this, Denver location, of course. We have the wildfire information, which is minimal being a metro location. Earth risk and previous seismic activity, historical weather risk with hail being the the highest kinda impact to this area, it looks like. Property fire risk and response times from fire station locations and detailed, crime index against state national average. It looks like crime is fairly high in this location, and then detailed, property information as well. Now this is just a select set of information that I I chose to return for this example. But the the power of this is I can run this against different locations across the country, different times of the day, and the answer stays in the same consistent format then I can then make trusted, business, data workflows and business decisions on with reliable, consistent results, for different locations. So same format, different results, obviously. So now if I, hop out and take a quick look at, what we're looking for moving forward to do is create a series of apps and tools based on APIs, trusted data, and MCP server, and have these AI integrations available on the platform. And just an example of how you can then use that the guardrails, and AI with our APIs together. Just a simple example of how that can be pulled into a AI developed, report here on the suite. For instance, an underwriting report that could be built with trusted data and MCP server. So here you have similar information pulled into a, preservable document. They'd be able to download or save, related to this property record and stored are for your organization. So like I said, a fairly quick and dirty demo, but just wanted to provide an intro to the MCP server conversation, think about the power of API, or sorry, AI apps and tools that can be built, Sure. with the framework of trusted data configuration and, We, AI, used together. And with that, I will we actually have a couple of questions for you, Dylan. One is the skill set that you showed in Claude, is is that required to use the EICP? it is not required, but it's recommended, to really the power of MCP server is, first, you can create actions. And when you ask a question to Claude, the the l m will search its available actions to answer, and you can combine those actions with web search or turn web search off. But if you develop skills on top of actions or projects, skills, and actions, you can give specific instruction to the LLM, to the agent on the order to query and answer a certain question and the specific tools to use and the method to use them. So you can just the skill refines your results even further, can improve performance and the reliability and certain certainty of your answers. Great. One more quick one is, how hard is it to configure the precisely MGP server? I don't believe it's very difficult, but it, you know, Well, that's my. if you use the Precisely MCP server, you just download it and install and and go, basically, if you're using, MCP with our APIs that that you're already using today or or may use in the future. So if you're using our geo addressing APIs and data graph, an additional suite capability that we're gonna be including in a unified, MCP server configuration, it'll be plug and play, essentially. Great. Thanks very. much. You install it. And I do want to note that the page that you showed for at the very end there, the precisely MCP server page, if you click on the docs icon next to the between the chat and the q and a button, The first link there is a link to that page, so you can go there directly. So that's there for any user on the on the on the call that would be interested in that. And with that, Dylan, thank you very much. We'll bring you back later on for additional questions. And, I'd like to invite Nina on the stage and go through the AI chat for, in the developer portal. Nina, if you would join me on stage, please. Thank you, Dylan. And it is my pleasure to introduce to you Nina Priyanka, doctor Nina Priyanka, who is in, India and will take us to the developer portal. Nina, go ahead. Yeah. Can you hear me very well? You are all set to go. Hear you very well. Alright. Thanks thanks, Dylan, for setting that stage onto the MCP server. That was a pop back presentation that you walked us through. And I'll talk about, intent, around the questions that the user asked about. How can we get that started quite quickly, and how easy is this to install the MCP server for the for for our APIs? So with that, I start, let me start with a confession. I'm a product manager at a data company, and my entire job is to make data easy to use. And yet, the last time I was to test to a new API, it took me two and a half hours just to set up the environment. Two and a half hours, not just to use the API, but just be allowed to talk to it. And here is the fun fact I would, like to quote here from a developer survey. 54% of the developers says poor documentation is the single biggest barrier to API adoption, not the technology, not the pricing, but the documentation. So, now that's the problem I take personally. Here is what traditionally happens when your team, whether you are in insurance, retail, bank banking, or telco, wants to evaluate new dataset. You schedule a meeting. You get a proposal. You get a sandbox. The sandbox take three weeks to provision. You get an API key. The API key doesn't work. You file a ticket. The ticket get resolved, and you finally run a query and the response schema doesn't match your data model. Six weeks later, you are back at square one, except now you have missed almost a quarter. We decided that was unacceptable based on the series of conversation and feedback we got from sales, engineer, customer support, professional services, and the client and partner like you itself. So we rebuild the entire developer experience, root, and branch to meet your needs. So today, we are talking about the developer portal to the the developer.cloud.precisely.com. It's a complete developer experience built around the six foundation things you actually need. APIs, 40 plus across 70 capability areas, like geo addressing, data enrichment, spatial analytics, data catalog, quality, integration, and just launch the, the new section that is the admin and the consumption, APIs like you where you can query the use the API and do your understanding around the consumption and the usage of your datasets. So that in real time, there are no build shocks that happens towards the end of the in tight in end of the subscription that has happened. MCPs. MCPs are directly available on our developer portal, which I which I think it's the talk of the town on every converse client conversation or every partner conversation we have. So your AI agent is live on on the connection for the precisely data, and we'll get there in a minute. Documentation. Real quick starters, which really understand the integration patterns that you are looking for. Authentication guides, the, API fundamentals. No theoretical documentation that exists. It's a proven documentation through which you can interact through the chatbot, and I'll be coming to that chatbot in a minute. Then there is a section that talks about the live demos that is try any API directly into the browser. Real data, it's a zero setup for you. You can query our, dataset through the interface of, say, ecommerce what our web website of ecommerce would look like, or a KYC customer would be filling up the form kind of arrangement that has been made into our that has been baked into our, demo portals. Release notes to stay updated with the change every deprecation that's documented and dated. So your inter the integration doesn't break into any surprise. And last but not the least is the chatbot and the MCP playground, which I'm going to talk in detail and show you in action today for for our discussion. So I would say all of it is available onto the, developer.precisely.com. Bookmark it now, and you will definitely use it every day for sure. So here is what traditionally happened, when a developer had a question about an API. They searched the docs. They get 17 results. They opened six of them. None answered the question. Then you go ahead and post into Slack and wait for a couple of hours to somebody to pick up that queries and get a response. Then someone sent you a link to a different doc, which had a broken ring. This entire broken conversation is what we have tried to fix through this DIS chatbot, which ends that loop entirely. You ask a question in plain English, and you get a direct answer, not a list of result, not a redirect, an answer that you really need. So these are the two dimensions that I'm going to explore with you and, make you aware how that how does it work in life with our, developer portal AI system that exist. So with that, I would like to share my screen. Yeah. Please let me know if you are able to see my screen. Yep. I can see your screen and I can see us on stage. Now you're on the developer portal, which, Fair the. way, in that same doc link that we had before, is the link to the developer portal where you can go directly. Yeah. So this is the, developer portal, that I was talking about. As you see, it has got the APIs, which has listing on all the APIs that I I already talked about, and you can explore it at your leisure. MCP takes you to the GitHub repository, which has the entire, documentation how to set up around it. Documentation as a it's a self explanatory. It's the documentation how for all the APIs that has been listed down here. Demos, basically, opens up the sections where you are able to explore our APIs on the on the run time. So it starts with at do addressing and address parser verification, and we'll be adding more and more onto the demo front. So how your front end would look like is is a kind of mimic that we have tried to do it here. So with that, let me, take some some of the queries around the developer portal. So this is what our developer assistant look like. It's onto the right section. If you click on it, there are two sets of interface that exist. One is the API interface, and one is the API playground. API, reference is basically as it is an interaction with our documentation, whereas API playground is nothing but an MCP powered system where you are exactly interacting with our precise APIs in the real time. So let me start with the use case around a common example would be the first and foremost example that I'll pick up from my slide itself that what are the first steps to make make my first API call. And if you query this, by the time it renders, I would walk you through what what does it entail and what information does it fetch it for you. So it tells you about these, how to sign up, to create the API key secret, the authentication method, the better token. So all the necessary details that would require to make your first API call successful is laid down here in the form of the documentation along with some quick checklist and the postman collection information as well. The second question, I would like to pick up that if I want to know what basically may my, API endpoint do. So this is the second question that I have already seeded, but you can do it live as well. Like, what does basically the address verify endpoint do? So it talks about the primary functions of the verify endpoint, the key capabilities. What's the endpoint that you you should use? And it also talks about the rate limit, the authentication information that you need. Third use case that I have picked up here is around an ecommerce checkout services. It's like asking the documentation, directly that what are the different sets of APIs that has required to make, ecommerce checkout service successful. And we have seen this implementation with couple of, other clients. They do make use of two two important, APIs that is the address auto complete for definitely, getting the real time address suggestions as the ship shipping or billing addresses is tight. And user, the company would also like to verify whether it's a deliverable or whether it's just a mailable address that we we that the client has typed so that there is no failed deliveries that occur at the end of the day because return logistic is a higher cost as you know. Besides the geo addressing component, I also wanted to talk about the special, APIs that exist within the developer portal as well, like routing, drive time. You can query those APIs and do your drive time or navigation related analysis in in a quick informative way. Many of us in fact, many of the customer do ask about the questions that what are the different request and response attributes that you have, and how can I get a CSV export or or a schema definition of it? So you do not have to come to us to ask those questions anymore. You it's there live. You can interact, and you can know what is the type of parameters that's being used under the request. The same applies for the what's the preference parameter that you want to pick up, the response fields that you're going to get at the end of the day out of the, API service that you are using. So the list is enormous in terms of what's the kind of questions that you are looking for and the type of response they are that you are looking for in terms of in terms of interactions with our API documentation. And before closing on the DIS chatbot, I would like to say our developer life is incomplete if we do not have an error to debug. So I took a question around why I'm getting a four zero one error when calling the API, and it tells you what are the things that could be missing, the common use cases, how to verify your API credentials. You can check about the, look at the quick troubleshooting checklist to know where you stand in terms of finding finding the information if you are stuck in debugging things. So in under ten seconds, I would say the DIS chatbot diagnosed it for you. And last one alone saves the developers three to four hours a week compounded across a team that is thousand of hours a year that go back in building and not debugging the product. So here's the number that should matter to you. Teams that onboarding using a conversional AI layer are producing 60% faster than those using traditional documentation. Not because the technology is different, because the experience is different once you know what you are looking for. With that, I will take a pause, on the if there are questions before moving on to the MCP developer MCP, use cases. If anybody has an additional question for Nina we had one question come in that we'll ask Dave when he joins, Sure. but, Nina, I think we can move on to your next slide. Can I switch gears from, just to give a little bit introduction onto the MCP server first and then showcase the demo? Yep. Let me launch that slide for you, Nina. Yeah. Slide 35. Yeah. Amanda, can you launch that slide, please, back in? Nina, I think you have to stop sharing first. Okay. Sorry. There you go. back. There you go. Now we're all set. So we go slide 35. Yep. Okay. Here you go. Yeah. So as as we know in the previous conversation with Dylan that there are MCP inaction, many of us wanted to use it live right away without waiting for the hassle of getting integration needs to happen or how do how do we install the MCP. So the DIS MCP playground helps you close that gap. So as you know, the MCP stands for model context protocol, and think of it as a live wire between your AI agent and our data. So no more data, no sandbox fake responses, real addresses, real schemas, and real intelligence right into your browser in just two minutes is what we have provisioned onto the developer portal. We connect cloud, cloud, and, and it's directly interacting with the live APIs, so you do not need a local setup at all. You do not need a credentials to configure to get started. All you need to do is log in to the, developer.cloud.precisely.com, open the browser, start querying it, and that's it. And the questions that you see here is the ones that, that, again, I'll be demonstrating in the screen shared with you. But I just wanted to let you know that these facts that you see on on my slides here, what is the coastal risk for, say, this particular address in Florida, or what is the crime index for 42 value of the Sand Drive Fair Play Colorado? So these intelligence are brought from our real time APIs just within the, fraction of seconds of you typing in the questions. The another, two use cases that I'll be going to show you live is also that how to return the full response schema for for a particular address. And it's not only related to you not only, restricted to US addresses only, but it goes beyond to be the international or global in true sense. So with that, I will share my screen to show showcase how the results are being fetched. Thank you, David. And let me know once you are able to see my screen. Yep. All set. Alright. See, the question that I see dated there, it's it's just a seed question for sure, but you can put any number of questions that you want to want you want this API play playground to explore. And how you can go here is the second tab that I was referring to. Sorry. Stops. So, this is the API playground where, the getting started queues are given, and it it covers up a whole, length and breadth of our analysis or around the API. But I'm going to show you a few use cases that I talked about. So if you look at what's the coastal risk for Hundred Ocean Drive, Miami Beach, Florida, It pulls the information around the location overview, the proximity to coast, the hurricane wind risk, which, Dylan already talked about, the risk summary that it entails. Second questions were around the crime index, and, again, it's able to pull the crime index summary. It gives you the index score compared to the national comparison and so forth. So you can play around with it as many of the questions or the kind of information that you are looking before buying our solution, for for real time implementation. So with that, I'm coming towards the end of my We we the have a question for you, Nina. Somebody asked, Yeah. can we can we ask a question like, is this a real address that is occupiable or a valid SAG address? Can you ask questions like that? Yeah. You can do ask questions around the, deliverability and existent of existence of that address for sure. So I don't think that should be a barrier around, in in terms of the precisely API conversation. that you can establish. With with that, again, we did put a link to the developer portal in the docs, section so you can, go try it out yourself. So, I want to say something, since I have taken most of the time, I would say. I want to say something directly to the customers and partners on this call. You did show up because you you didn't need another slide deck for sure. You showed up because you are really building something, and you want the data behind it to be right. So one more start before I go. The same study that found the documentation is the number one barrier. The adoption rate rates definitely went up by more than 40% if you follow the, ground rule of exploration and learning first before diving it into buying into the buy me persona. So it's an, the DIS chatbot and DIS MCP playground are like they are not products. They are an invitation. An invitation to ask your hardest questions about your address, customer, your risk exposure, and get a real answer live before you haven't you spend a penny with Precisely. So with that, I will hand over to David. Thank you so much. for listening you, to me. Nina. Dave Mosher is gonna join me next, and we're gonna talk about notes from this field and shared some stories. And most of you, I believe, know Dave. And, Dave, over to you. Thank you, Nina. Awesome. Thank, that was. fantastic, Nina. And and, Dylan, fantastic work too during your demo as well. So you set the stage perfectly for me. But but before I get into that, David, I'm not sure if you feel the same way, but there can never be enough Davids on a phone call. So the more Davids added to this, you know, moving down the line, absolutely. Add add as many as we can. Right? But, anyways, that being said, what David had asked me to do is join the conversation or join the, join the group and share some use cases that I've been discussing in the field with current clients, both past and, obviously, current activities and or opportunities I'm working on clients with. And so but like Nina and Dylan, I wanted to set the stage a little bit, make sure we cover off on the MCP and close the loop on that before we get into how we're seeing it being utilized in more of a real time fashion or production based fashion by our clients as well. And, David, I I think there were a couple questions in the chat. Did you want me to address those now before we get into it? Yeah. There was one other chat, Dave. Does precisely have an exposed MCP server that consumers can directly access? Not one right now that we host on behalf of our clients. They'll the clients our clients will have to install the MCP, either on their local PC and or a server of their choosing to access our MCP. But that is coming shortly. So, again, Dylan and I'm sure Nina will speak to that during the next forum, but not right now from a hosted MCP perspective. Great. And one other one is, does the MCP server work for non location intelligence and data enrichment use cases? It does. Absolutely. We've actually seen a number of our clients starting to customize the MCP. So actually add layers as a part of the out of the box MCP that we deliver or have a get available for and put in things like our data quality APIs, because what we're seeing is, of course, the distribution of the, authorization information across the APIs is contained and managed by the MCP. And what that affords our end users to do is start exposing and or gaining access to those other APIs in the portfolio outside of location as well as enrich, APIs into data quality and other factors like even data catalog at times. So short answer, yes. Long answer, definitely give us a call. We can absolutely coach you through that. Great. Thank you. Yeah. Of course. Take it away. Cool. Cool. So I'm gonna flip ahead to the next slide here. Actually, one more slide. So, again, just to kind of round up the conversation we've been having about the MCP. Obviously, Obviously, that was the the core of our conversation today because a lot of people a lot of our clients are asking questions about it. Two things I wanted to clarify, and I believe Dylan and Nina both touched on this, but it doesn't have to be a chat based interaction for you to leverage the MCP. An MCP essentially is, as Dylan had mentioned, great analogy for this, guardrails around your AI. You can let an agent interact with this MCP in an autonomous fashion as a part of your workflows and let it just scream and leverage that trusted data from precisely albeit via APIs like we had talked about today or just bulk datasets that you have available in other data platforms, which we'll get into here in a second. But the vast majority of our clients, those that are looking to get more value out of their AI spend, are looking to investigate or integrate, excuse me, our MCP directly within those chat instances. Right? The chat assistance. That could be Claude. That could be chat GPT. That could be, you name it, insert whatever. Right? Copilot, so on and so forth. With that, essentially, it's bringing trusted data, to the end users immediately. Right? A lot of our clients on top of this too are starting to expose their internal datasets, precisely datasets, to their internal users through the same type of interaction of an MCP that they've built. What we wanted to do was provide our clients the ability to interact with our APIs directly through an MCP that we've created with defined protocols in place. Actually, guardrails, like Dylan had mentioned, but it's the art of the possible when it comes to these things. There's a total of, I think, roughly about thirteen thirteen thousand MCPs that are supported, you know, by other vendors across the marketplace today. What's really fascinating is during conversations we're having, with our clients is them asking questions like, can you start stringing these MCPs together? Right? Can I start using potential data integration MCPs along with your MCPs? So as data's moving from point a to point b, can I enrich it with some additional context directly through this series of protocols? And that's essentially what builds into that skills kinda conversation. I believe that was the question that was asked earlier too. So the stringing of MCPs is certainly something that we're seeing a lot of our clients start to do. And then again, it can be directly through a chat based interaction or it can be through an agentic workflow, that you're deploying in a number of different data platforms. So I was actually on the phone with a client last week who's actually a heavy, enriched user of ours. So they leverage all of our property datasets. All of those are exposed through the share directly within Snowflake that we support to them with them. But what they also wanted to do was essentially set up an MCP interaction through Cortex instead. So, again, just because we offer up an MCP doesn't mean others aren't using MCPs within those platforms and potentially leaning on, in this case, calling out to some of the other APIs that maybe they don't have access to, like the risk attribution to enrich in kind of ad hoc based queries outside of the datasets that they essentially subscribe to, with us in bulk. Same exact things coming into play here on the Databricks front with Genie based interactions. Right? Creating notebooks and or urgent workflows or lakeflows as a part of their plat as a part of the platform and essentially setting up agents to call out to these MCPs in a step by step fashion based upon the protocol you have in place. So, again, when we're talking about MCP, MCP is not a precisely specific terminology. Right? As all of you are, I'm sure, aware, there's a ton of them out there. There's a ton of things that people are doing with them, and they're creating them left and right too. So I have on my PC alone about, I wanna say, about a 150 MCPs, my own personal MCPs that I've created for different purposes, right, that I'm leveraging Cloud, but also our internal Copilot, you know, to interact with. And on top of that, also starting to incorporate interactions with our flagship products too or our core products in the sense of, like, Spectrum users or, you know, for Trillium based interactions. There's a lot of possibility and potential when it comes to these MCPs, but at the end of the day, really, what we're trying to do and facilitate is trusted responses more than anything else. Right? You're not gonna get the most out of your AI spend unless you trust the answers and the responses that your AI is providing you. So couple of kinda high level examples of that. And, again, just a plug here too to the audience. This session was actually set up, notes from the field, to be both me just talking, but I'm sure people get tired of hearing me talk. So I would love to have a client or anybody on the phone join me on stage next time for this session to talk about a use case, right, in which they're leveraging our data integrity suite APIs or broader data integrity suite portfolio. Would really love the opportunity to speak to you and learn a little bit more from you, and I'm sure that other people on the phone, would certainly like to hear from you as well. So I'll pause. Are there any other questions, David, that came up from the chat? I don't see any, questions, but certainly, you drop it in the q and a or in the chat or if anybody has a question that can raise their hand and join us on stage, we would love to have you do that while we sort of wrap up the end of our time. We are, just a mental note on time. We're at the top of the hour. We do have a little additional time. We have a couple of, additional things. So if you can hang out with us, that'd be great. We, sorry for running over a little bit, but we, but we have a little bit more content. If you wanna stay, feel free to stay. That's, one of the get that out there. Cool. And and I guess the one thing the one one additional thing I wanted to say too, David, is, again, I know we're talking about API interactions, right, through the MCP. Again, an MCP can be set up to query against anything. Right? In this case, you can see the actionable insights at the tail end of this. Even though our APIs fit in very well into the framework of this, that's essentially what we built out first to make it easier to interact with the APIs because as Nina had alluded to, it's difficult to get up and running on them. So let's make it easier for our clients to leverage them. But this can be against datasets, databases. I mean, you name it. Right? It's the art of the possible when it comes to these MCPs. So, again, think about different ways in which or MCPs that you would like us to potentially build out on behalf of you, for you to begin ingesting into your agentic workflows or any of your AI based use cases and initiatives. So but, again, appreciate everybody joining today, and happy to hop back on during the next forum and give you some a little bit more of a download on what we're hearing from the field as well. Thank you. Great. Thanks, Dave. Antonio, if you wanna join me back on stage Dave, if you wanna stay there, that'd be great too, Sure. Yeah. as we sort of, kinda conclude our thoughts and have any, additional wrap up, comments. Antonio, over to you. Yeah. So at this point, this was when we wanted to kinda hear from from all of you, in, in the session. If there were any specific questions or if there were, discussion topics that you all wanted to cover, it's it's important that we continue through our conversations and continue to build this community. So, as as David had mentioned, if there's anybody that would like to kinda speak up or any questions that you that you have or comments on on or feedback on the material, we'd love to hear from you. Yeah. And while we wait there for a second, Amanda, if you could launch the fourth poll, please. As we wrap up, we are, starting to plan, believe it or not, tomorrow, our next, group. So we do have a couple of question, couple options. What would you like to, cover next in the next group? There's a list of other, qualities there that we can talk about. So as we sort of wrap up with any additional questions, if you'd answer that poll, that'd be great. We'll give it a minute here. Looks like we're right on the ball with AI data and location intelligence. Yeah. Geo addressing, spatial. Great. There's some great insights there. Well, with no other questions, why don't we start to to wrap up here? Amanda, if you could stop sharing the poll, but keep it open. Feel free to answer it as we go. So, we do have a well, no. That didn't work work exactly as I thought it was gonna be. So let's get to the wrap up here. Couple of notes as we wrap up. We are launching a data integrity suite, knowledge community, that Nina is running on the back end. So there's a brand new community, grow your knowledge and grow your network. Feel free to join us. There's a link in the doc section here. Ask questions, give answers, interact with your peers and with subject matter experts. We'd love to get you to know you on the knowledge community. It's brand new and it will be growing over time. My colleague, Matt has been, adding a lot of great content and we'll continue growing that over, over the year. We'd also like if the Dave said earlier, if you'd like to share your expertise and feedback and share your story, we'd love to have you join us on our next event. We're looking toward the June, so it'll be a quick turnaround time. If you've, got some expertise or made an impact at your company with anything that we talked about or using the data integrity suite in combination or your experience with MCP, we'd love to hear your story. You can email me at david. Leahprecisely dot com, and we'll connect you to the right folks for that. We're also excited to announce that, Precisely has some brand new support plans, standard, advanced, and enterprise support that are rolling out across the company, and they're brand new to offer you more value and offer you deeper service. And as I said, right before, watch for our updates for the q three, suite customer forum. We will be, launching that in, very shortly. You should see an invite for that. So, thank you for your time today. On behalf of all of Precisely, we'd like to thank all the customers for attending. We can't do what we do without you, so we appreciate you and value your input, as we roll out and make additional enhancements to the suite. Antonio, Dave, any last comments before we sign off? I think the biggest thing for me is that I really appreciate everybody showing up here. And and, even even if we didn't get everybody on on stage just yet, I know it's gonna take a little a little bit of time, but I had some awesome engagement and some great questions in in the chat. So just thank you all. This is this wouldn't be possible without you, you being here and and showing up and being part of this. I I 100% echo what Antonio said. Thank you all for joining today. Thank you for being precisely customers, but also valued partners of our organization as well. And as David mentioned too these are going to be coming out fast but technology is moving fast right in the world of AI and this day and age of AI we just need to continue moving fast and we're here to help support you along your along obviously, along your AI journey overall. So, again, thanks everybody for joining today. Really appreciate it. Looking forward to the next one. Thank you, and have a great day, everybody.